I've been having this problem for a couple months now, but today I'm annoyed enough to ask if there's something I can do about it.
Normally, I run esync as a cron job in the middle of the night (the standard 3AM), and normally it works fine (though I had a few days where the error I'm detailing below also occurred, but that seems to have cleared up). I get the mail and it has information as it should. It even went fine this afternoon, as my computer was off at 3AM, due to severe thunderstorms, so the cron job ran at 1:20 PM when I turned it back on. However the second cron job was for revdep-rebuild -p, which suggested that I'd want to rebuild hal and dbus (not expected, but not unreasonable based on my activities of the last couple days), and when I tried to do that, I got the dreaded !!! Security Violation: A file exists that is not in the manifest. error. Which was bizarre, but no need to freak out; I know that that such an error will likely be fixed with a sync. So I tried to esync with my usual mirror pool, rsync.europe.gentoo.org, but, as seems to happen if I sync in the afternoon (the previous sync not withstanding), it failed. The sync started, and progressed quite far, but then-- (errors from running emerge sync in a terminal, so I could see the output, having no patience to read /var/log/emerge-sync.log for errors): io timeout after 90 seconds - exiting rsync error: timeout in data send/receive (code 30) at io.c(109) rsync: connection unexpectedly closed (2970632 bytes read so far) rsync error: error in rsync protocol data stream (code 12) at io.c(189) |>>> retry ... receiving file list ... 124787 files to consider io timeout after 90 seconds - exiting rsync error: timeout in data send/receive (code 30) at io.c(109) rsync: connection unexpectedly closed (3049937 bytes read so far) rsync error: error in rsync protocol data stream (code 12) at io.c(189) All three retries, using mirrors from the Europe pool in Germany, Finland and The Netherlands failed to complete with these errors. So, as I always do, I changed my rsync mirror in /etc/make.conf to rsync.us.gentoo.org, and it synced perfectly on the first try, as it always does when I encounter this issue (and being out-of-sync *was*, of course, the problem with revdep-rebuild, which is now running its oneshots with no complaints). What is going wrong here? It can't be my implementation of rsync (latest stable), or the US pool would fail as well. It could be the servers in the European pool, but... all of them? Or am I simply using the wrong SYNC= in /etc/make.conf? My default is rsync.europe.gentoo.org, and I don't see any documentation that indicates that that has changed or become deprecated or invalid, but maybe it has. I don't particularly want to switch permanently to the US pool, as that just seems to make more vectors of instability for everybody, and is not really the point of having continental mirror pools anyway. Anybody got a clue as to what's happening and what, if anything, I can do to fix it? Holly -- gentoo-user@gentoo.org mailing list
